Bringing TBI and Multi Port Fuel Injection to a New Level.     EFI Conversions and Tuning! Seattle to Portland! E-mail Tuning Consultant!
Results 1 to 15 of 17

Thread: Open Loop EE Advice

Hybrid View

  1. #1
    Fuel Injected! spfautsch's Avatar
    Join Date
    Apr 2015
    Posts
    53

    Open Loop EE Advice

    I've been struggling with knock on a freshly built LT1 for several weeks. It seems to be happening in taller gear ranges at cruising speeds, and only became blatantly evident when IATs started staying above 80f all day (would see little or no knock in the mornings when IAT was lower).

    Engine Specs:
    Stock crank, new scat rods, cast .030 over pistons, recip assy balanced within .5 grams
    10207643 casting heads, LE1 port job, manley 1.94 / 1.50 ss valves, Lunati 73925 springs
    LE ported stock intake
    CompCams 07-467-8 - 230/236 @ 0.050", .576 / .570 lift, 113deg LSA installed 3 deg advanced for 109deg ICL
    CompCams 1605-16 Ultra Pro Mag 1.6 RRs, TFS guideplates, ARP 7/16 studs
    Remanned stock opti w/ OE spec pickup, new coil, ICM, 8mm wires
    StainlessWorks 1-5/8" standard length headers w/ high flow cats into 2-1/2" duals w/ X pipe

    Compression ratio should be more or less stock. Other than EGR delete / block-off all other emissions devices are functional (AIR, CCP).

    I've tuned VE using narrowband data with the assumption my closed loop calibration was more or less in the ballpark of accurate. That said, a large amount of VE was removed below 2500RPM - in the neighborhood of 25%. I'm sure some of this is related to the cam, but as the IATs have been climbing I've noticed ever increasing knock in these cruising ranges especially in higher gears, to the point it's stumbling so bad I thought I had a wheel bearing about to lock up. Having pulled an enormous amount of timing out of this area with no result, I'm relatively sure it's going lean and needs more fuel due to the reduced effective displacement caused by valve overlap. I've chased down all the usual causes of false knock and feel relatively confident most of it's real. Plugs show no signs of severe detonation but are otherwise bone dry. Above 3000 RPM or in PE (still using stock PE AFRs) it runs like a top and no knock. So this morning I used eehack to disable BLMs and command AFR to 13.5. Knock isn't completely gone but greatly diminished, and when I turn BLMs back on the knock count starts rolling again.

    Even though it idles great in CL, I think I want to try running OL with MAF and wondering if there are any pitfalls I should be aware of. Am I safe leaving the cats on?

  2. #2
    Fuel Injected!
    Join Date
    Oct 2013
    Posts
    728
    I had to turn-off all the knock retard/detection tables. I was constantly getting knock retard even though it wasn't knocking. It seems that more lift and roller rockers mess with the knock sensor unless you have a LT4 knock module.

  3. #3
    Fuel Injected! spfautsch's Avatar
    Join Date
    Apr 2015
    Posts
    53
    I've researched up and down that subject and though I'd be open to trying one out, LT4 knock modules have become as rare as hen's teeth. Without a wideband and better knowledge of what I'm doing I'm extremely reluctant to detune knock retard.

    I'm sure some amount of what I'm logging is false, as I can get the knock count to increment at hot idle by just blipping the throttle. In addition to that, the exhaust system seems like it resonates well at the frequency the knock circuit is tuned to. I picked up huge amounts of knock count by whacking the collector with a small piece of brass rod. I also found I had cracked one of my exhaust tips with a backfire (killed my MAF cleaning it, severe flood condition) and it's "quacking" like a goose was picked up. I've also seen drivetrain noise set it off, hard shifting, acceleration over rough pavement.

    But...

    The really debilitating stuff is happening between 1300-2100 rpm at 60-75 mph. In cold air it's barely noticeable, but the warmer and thinner the air gets, the worse it becomes to the point it "feeds" itself and KR ends up pegged at 12 until you downshift. Switching from OL to CL with eehack, the counts start climbing as soon as BLMs are enabled so it seems pretty obvious it's going lean in CL.

    Knock graph from 5-12

    Current OL tune
    Attached Images Attached Images

  4. #4
    Fuel Injected! Terminal_Crazy's Avatar
    Join Date
    Oct 2015
    Location
    Lancashire England
    Posts
    122
    My take on this as i've been through it, is most of what you have read about anything doesn't relate to any observations you will see.
    Most of my changes have seemed counter intuiitive to anything i've read. YMMV

    Steveo's Trimalyzer is now on version 1.3b
    Have you viewed the knock counts in EEHack and in the analyzer view?

    The LT1 doesn't have AE acceleration enrichment so any blip of the throtlle leans the motor until the PCM can figure out a new pulse width.
    This is/can be burst knock which has it's own tables of retard.
    You can get knock without counts increasing & increasing counts without knock.

    OL runs with a lot less counts as i can run it richer. In CL i keep retarding timing and still get knock counts.
    That image does look heavy though. So to cover some ideas, You have checked nothing is loose or catching underneath?
    You have the right knock sensor for the PCM?

    Also my current PCM has a fewer knock counts with an LT1 knock module than the previous PCM with an LT4 knock module, so don't get fixated by that notion either.

    HTH some
    Mitch
    '95 Z28 M6 -Just the odd mod.
    '80 350 A3 C3 Corvette - recent addition.

  5. #5
    Fuel Injected! spfautsch's Avatar
    Join Date
    Apr 2015
    Posts
    53
    Quote Originally Posted by Terminal_Crazy View Post
    Steveo's Trimalyzer is now on version 1.3b
    I'm aware. I've modified the build I'm running to ignore frames with corrupt data as these inevitably get picked up as knock events. See here. Some of my logs have as many as 300 trashed frames, but my commute is about 65 minutes each way so that's a lot of data. I think this might be a y-body quirk but haven't had the time to investigate further.

    Quote Originally Posted by Terminal_Crazy View Post
    Have you viewed the knock counts in EEHack and in the analyzer view?
    Always.

    Quote Originally Posted by Terminal_Crazy View Post
    This is/can be burst knock which has it's own tables of retard.
    It was my understanding that burst knock is pre-emptive timing retard that is added to compensate for the lack of a pump shot. I would have to review all the code again, but as I recall all of steve's knock analysis routines only plot an event if knock count increments [edit] regardless of if there's knock retard.

    Quote Originally Posted by Terminal_Crazy View Post
    You can get knock without counts increasing & increasing counts without knock.
    Did you possibly mean "increasing counts without retard" in the last sentence? If so I think what you're getting at is knock in MAP cells lower than 45kpa generates no KR. Got it. [edit] If I cover up the knock plot with my thumb so I can't see anything to the left of 45kpa I'm no less freaked out.

    Quote Originally Posted by Terminal_Crazy View Post
    OL runs with a lot less counts as i can run it richer. In CL i keep retarding timing and still get knock counts.
    That image does look heavy though. So to cover some ideas, You have checked nothing is loose or catching underneath?
    I'm going to get it on a lift tomorrow and verify, but yes. I see nothing to indicate the exhaust is touching anything. I'm going to put a fuel pressure gauge on it tomorrow too - I installed a new racetronix 255l/hr pump before I started doing any serious tuning and pressures were good after that so I didn't think about it further.

    Quote Originally Posted by Terminal_Crazy View Post
    You have the right knock sensor for the PCM?
    Pretty sure. Delco 10456126 both replaced during build.

    Logging tonight in OL wasn't as good as I'd hoped - the knock plot isn't great (456 points in ~60 miles). But I still feel like it's running leaner than I'd like b/c I filled up and calculated 22.3 mpg for today. Seems unlikely with the way I drive. But some logs from last week were in the neighborhood of 1100 events.

    I feel like some of what's getting picked up as knock may be misfires / lugging that gets amplified by the drivetrain. It's a 22 year old car and has rattles and squeaks like any other with 160,000 mi on the odometer. Maybe I'm operating it in a load range that the cam won't allow except with dense air.

    Thanks for bouncing the ideas around - always good to discuss with someone who's been there and done that!

  6. #6
    Fuel Injected! steveo's Avatar
    Join Date
    Aug 2013
    Posts
    2,166
    assuming you have ruled out a dead fuel pump or large fueling error, just disable the knock sensor already. you really don't need it if it's causing you this much trouble. knock detection on modified engines is hit or miss, yours is a miss.

Similar Threads

  1. Replies: 3
    Last Post: 09-06-2016, 07:06 AM
  2. Open loop afr
    By myburb in forum GM EFI Systems
    Replies: 10
    Last Post: 07-13-2016, 07:05 PM
  3. Open to closed loop
    By myburb in forum GM EFI Systems
    Replies: 0
    Last Post: 05-20-2016, 01:43 PM
  4. Staying Open Loop
    By rsicard in forum TunerPro Tuning Talk
    Replies: 7
    Last Post: 03-18-2016, 11:47 AM
  5. EE open loop idle
    By steveo in forum GM EFI Systems
    Replies: 17
    Last Post: 01-27-2014, 05:41 PM

Bookmarks

Posting Permissions

  • You may not post new threads
  • You may not post replies
  • You may not post attachments
  • You may not edit your posts
  •